high doppler resolution x-band wake-vortex radar : cdg airport

Air Systems Division

High Doppler Resolution X-bandWake-Vortex Radar : CDG Airport Trials F. BARBARESCO


Brieve History of Wake Vortex Radar Campaigns

Air Systems Division3

GEC MARCONI EXPERIMENT (UK) : 1992GEC-MARCONI DX 04 Radar (S Band)Wake Vortex have been clearly detected (measured RCS : -80 dBsm to –90 dBsm) at 2.8 Km by radar (radar beam perpendicular to the aircraft’s flight path)DX Radar Figures :3 GHz (λ=9 cm)PRI = 100 μsτc= 4 μs (B=0.25 MHz, 600 m) to 13 ns (B=77 MHz, 2 m)τ< 400 . τcPc = 10 KWτ/PRI < 2.5 %GEmGRec= 33 dB

Radar Beam Position

HS-748 Plane :

Width : 30.02 m

Weight : 40000 LBS

Speed : 138 kts (69 m/s)

Weather conditions :

Cloud base : 2500 feet

Humidity : 85%Doppler Spectrum(range cells : 15 m)


CNRS/CRPE, CNET & STNA (France) : 1992PROUST RadarFrequency : 961 MHzWavelength : 31.2 cmPRI = 156.4 μsτ = 1 μsPc = 4.5 KWResrange = 150 m (withoutpulse compression)τ/PRI < 2.5 %Beamwidth : 5°Gant = 29 dBRangeamb = 23.4 Km

Doppler Spectrum(range cells : 30 m)


CLEAR AIR RADAR REFLECTIVITY OF WAKE VORTEXTests have revealed radar echoes in clear air.

Two mechanisms causing refractive index gradients are :Radial Pressure (and therefore density) gradient in a columnar vortex arising from the rotational flow :

Adiabatic transport of atmospheric fluid within a descending oval surrounding a vortex pair :

Particulates were not involved (not f4 Rayleigh scattering) .The frequency dependence was not the Kolmogorov f1/3

The role of Engine Exhaust :RCS doesn’t change when the engine run at idle or full powerExhaust diameter yields a partial pressure of vapour and a contribution which is much smaller than that due to temperature.

⎪⎩

⎪⎨

⎧

=

⎟⎠⎞

⎜⎝⎛+⎟

⎠⎞

⎜⎝⎛+⎟

⎠⎞

⎜⎝⎛=−

∞ K(K) , Tmperature T : the tepourf water vaure (mb) otial press : the par; Pf dry air ure (mb) otial press : the parP

GHzbelow equencies air for fr of humid tive indexn : refracwith

TP..

TP.

TP.).(n

va

vva

288

20

107763864677101 256

[ ]

⎪⎩

⎪⎨

⎧

+===

⎥⎦

⎤⎢⎣

⎡⎟⎟⎠

⎞⎜⎜⎝

⎛++−=−

−

va

--

zsat

PPe, and Pnd AltitudΔz : Desceer) (in Ws.-s.r) N (in Summes.el : Nat Sea Lev

ter)ion parametratificatequency (sVäisälä FrN : Brünt-with

(z)T...

(z)P)(T(z)PRHΔz

g(z)Nρ.(z)n(z)n

int0300200140

1049377622310~

111

626


THALES RADAR WAKE VORTEX PROCESSING CHAIN


CHARACTERISTICS OF THALES X-BAND BOR-A550 RADAR

Main FiguresFrequency : X Band (9.6 GHz for ORLY Campaign)Minimal Range : 400 mMaximal Range : 40 KmRange Resolution : 40 mMechanical Tilt : +/- 24°Beam Width (elevation/azimut): 4°/2.7°Radial Velocity Range : +/- 26 m/sDoppler Resolution (& High Resolution) : 0.2 m/s

(0.04m/s)Mechanical Scan Rate : 8°/s (sur 45° pour ORLY)Peak transmit power: 75 WData Link : RS232 (x2), RS432 (x2), EthernetRecording System : All range cellsexternal link Ethernet 20 Gb/s

PRF = 3348 HzF=9.6 GHz

Vamb = λ.PRF/2Vamb = 52.31 m/s

(+/-26 m/s)

λ=c/F=3.125 cm


Wake Vortex Detection in All weather Conditions

No Rain

Minimum Temperature 8 °

Maximum Temperature 14 °

Rain Rate 0 mmWind Speed 22 km/hWind Direction : South

Rain

Minimum Temperature 9 °

Maximum Temperature 12 °

Rain Rate 1.19 mmWind Speed 20 km/h

Wind Direction South-East

RCS of Medium Aircraft Wake Vortex : 0.01 m2 S/N ≈ 15 dB (Range = 600 m)

NO RAIN RAIN


Speed Variance of Rain can measure

EDR & TKE(air turbulence)

Time (s)

0 m/s

+/-26 m/s

0 m/s

Tangential SpeedVersus Radius

SpiralGeometry

Cross-Wind Speed

⎟⎠⎞

⎜⎝⎛ +=⇒=⇒=

VVbbr

ddraer rb δ

πθθ 1log

21

WAKE VORTEX PROFILING : RADAR DOPPLER ANALYSIS


WAKE VORTEX PROFILING : WAKE VORTEX AGEPositive

Time/Dopplerslopes

ZeroTime/Doppler

slopes

NegativeTime/Doppler

slopesLow speedNegative

Time/Dopplerslopes


Wake Vortex Circulation Computation

[ ] 323 2. /

i )S(Vmomentndk=Γ

[ ]

[ ]∫

∫∝Γ

max

min

max

min

3/2

3/22

)(

)(2

V

Vii

V

Viii

dVVS

dVVSV

Wake Vortex Doppler Frequencies Extraction(Pre-Processing : CFAR on Doppler axis)


Wake-Vortex Detection based on Doppler Entropy criteriaIn the framework of Affine Information Geometry, Kähler metric is given by Hessian of Entropy, considered as Kähler Potential Function :

Entropy of Multivariate Gaussian Law of Zero Mean :

Entropy can be parametrized by reflection coefficients of complex AR model

( ) ( ) ( )enRRΦ πlogdetlog~ −−=

-RHΗΗΦg

jiij =

∂∂∂

≡ et ~2

[ ] [ ]0

1

1

2 ..ln.1ln).(~ Penkn)(RΦn

kkn πμ +−−= ∑

−

=

[ ]∑

=

−

−−

−

==

−=n

kk

nnn

zn

P1

20

10

11

21

1 avec

.1

α

αμα ( ) [ ]∏∏−

=

−−−

=

− −==1

1

20

1

0

1 1detn

k

kn

kn

n

kknR μαα

[ ]

( )( )[ ] nn

nnnnn

RRTrn

nnn

RRE

mZmZR

eRRZp nn

=

−−=

=+

−−− −

ˆet

.ˆwith

..)()/(1.ˆ1π

Doppler Spectrum

shape

Doppler Spectrum

Power


Thales Advanced Processing Chain: 3 Patents

Implemented in C language :3 times faster

than real time by“off-line” tests on 4 Threadswith Quadri-

Core PC

PRF = 3348 Hz

Implemented in Matlablanguage


“Off-Line” Tests on 4-threads quadri-Core PCC-Code Code has been parallelizedon 4 threads by divided 31 Range cells in 4 Domains (1.2 km)Each Range Domain is processed on 1 threadRange Cell resolution : 40 m

C-Code implemented on 4 Threads is 3 time faster

than real time :

1 second processed in 300 ms


Real-Time Implementation of THALES “Wake Vortex”Algorithms (e.g. : General-Purpose computing on Graphics Processing Units)

In Progress : Real-Time THALES Processing Chain

General PurposeGPU

Ethernet 20Gb/sDATLINK

THALES Processing chain (3 patents)THALES radar

HMI

PRF = 3348 Hz

Multi-CorePC


LIDAR & RADAR WAKE VORTEX DOPPLER SPECTRUM

AngularRadar resolution

Doppler radarsignature

Doppler Lidarsignature

LIDAR WAKE VORTEX DETECTION

Post-processingNeeded for Wind

Inversion detection


Paris Airport Radar Trials


Radar Sensor Campaigns funded by THALES

2006 : Paris Orly AirportFrom Mode S site (500 m from runway)Wake Vortex Monitoring during departuresWeather conditions : Winter (clear air, rain)

2007 : Paris Orly AirportFrom Thales Limours Testbed TowerWake Vortex Monitoring during arrivals (ILS Interception)Weather Conditions : Autumn (highly turbulent atmosphere)

2008 : Paris CDG AirportFrom Kbis Tower co-localized with Eurocontrol Lidar (2 mm Windtracer)Wake Vortex Monitoring during arrivals/departures on Closely Spaced Parallel runwaysWeather Conditions : Summer (very hot, rain)

Orly 2006

Orly 2007

CDG 2008


BOR-A550 Radar View at Paris ORLY Airport Site


PARIS ORLY WAKE VORTEX RADAR CAMPAIGN (2006)


Orly Airport 2006 (runway monitoring for departure)


Orly Airport 2007 (ILS Interception monitoring for arrival)


Wake Vortex Monitoring based on HR Doppler

Vertical ScanningFrom Limours Tesbed

Tower (Orly ILS Interception : 1500 m)

Highly Turbulent Atmosphere !!

Wake VortexRollups Wake Vortex

Rollups

Color = Doppler Entropy E = - log det(R) (R : covariance matrice)


CDG Airport 2008 (runway monitoring : departure/arrival)

08L

08R

26L

26R

27L

27R

HorizontalScanning

VerticalScanning


Scanning strategies

Vertical Scanning

Horizontal Scanning

360° scanning


Vertical Scanning: Monitoring of Arrival (2nd runway)

Wake Vortex Roll-up(Arrival)

DepartureArrival

East Configuration


Vertical Scan: Monitoring of Departure (1rst runway)

Wake Vortex Roll-up(Departure)

DepartureArrival

West Configuration


West Config. Procedure

From scan to scan : Departure


Vertical Scanning


Horizontal scanning


Wake Vortex Position & Circulation (strength in m2/s) along runwayPosition des vortex au cours du temps

1200

1300

1400

1500

1600

1700

1 2 3 4 5 6 7

Numéro de balayage

Cas

e di

stan

ce

vortex 1 position

vortex 2 position

Circulation des vortex au cours du temps en m²/s

0

20

40

60

80

100

120

140

160

180

1 2 3 4 5 6 7

N uméro d e b alayag e

vortex1 circulat ionvortex2 circulat ion

5 s per scan (35 s)Position at 1500 m

Position des vortex au cours du temps

600

700800

9001000

11001200

1300

1 2 3 4 5 6

Numéro de balayage

Cas

e di

stan

ce

vortex 1 position

vortex 2 position

Circulation des vortex au cours du temps en m²/s

0

50

100

150

200

250

300

350

1 2 3 4 5 6

N uméro d e balayage

vortex1 circulat ionvortex2 circulat ion

5 s per scan (30 s)Position at 1000 m

With Cross-Wind Without Cross-Wind


Thales Radar Trials SynthesisParis ORLY AIRPORT Campaign has proved that X-band Radar can :

Detect Wake Vortex (RCS of 0.01 m2 on medium aircraft) :In all weather conditions (wet & dry at short range < 2 Km)In Staring & Scanning ModeIn real Time / Update Rate of 11 s on 60° scan sector

Localize Wake Vortexin range/azimuthWith resolution : 40 m x 1°

Characterize Wake VortexGeometry (Spiral)Age (Young, mature, old, decaying)Strength : Circulation (m2/s)

Characterize ambient air (in Rain Conditions)Wind (based on Doppler & post-processing)Ambient Air Turbulence (EDR & TKE)


X-band Radar PerformancesWake Vortex Monitoring in All Weather Conditions (light to heavy

rain, fog, turbulent atmosphere,…)Operational Use to track Wake Vortex (transport, decay, rebound) in extreme weather conditions :

Wind burst (wind under Cb, turbulent atmosphere, wind shear…)No Wind (foggy weather,…)

Non Operational Use for « Safety Case » by Data Collection :Risks Assessment according to extreme wind conditions and not only on mean wind conditions (extreme values statistics)Need for Wake Vortex Data in exhaustive cases of airport climatology(good/bad weather)

Fast Monitoring of very large volume (radar scanning) with high update rate (e.g. : 8°/s with mechanical scanning of BOR-A radar and faster with Electronic scanning)

Highly sensitivity of Radar : monitoring of Wake Vortex for « medium » (high % of A320) aircrafts and not only « heavy / super heavy » (requested for traffic mix of « very light jets »)


Radar/Lidar Sensors

Radar & Lidar are complementary sensors in all weather operationsTHALES has proved by derisking campaign (Paris CDG-2008) that X-band Radar & Lidar are complementary for Wake Vortex Monitoring :

Lockheed Martin has proved by derisking Campaign (Westheimer Aiport) that X-band & Lidar are compementary for Wind Monitoring (& Wind-shear)


WAKE-VORTEX & WIND SENSORS BENCHMARK


Suite of Sensors benchmarking THALES will supervise benchmarking of :

Wake Vortex Monitoring Sensors

Wind Monitoring Sensors

Others

In All Weather Conditions

EUROCONTROL

M-Scan & E-scan X-band Radars

2 μm Lidar

Polar X-band Radar

1.5 μm LidarWind Profiler

UHF Radar Wind Profiler

Ultrasonic Anemometers

Sodar


Threshold

Touch-Down Area

Runway 3000 m / 60 m (typically)

Wind

Final Approach Fix (FAF)

ILS Interception Area

Altitude between 2000 and 6000 ft (typically)

H = 400 ft

400 m2500 m

25 000 m (13.6 NM)

ILS Glide Slope

ILS Glide Slope : 3° (5.2%) – 4°

Max course deviation : 2° Ie horizontal deviation at FAF : +/- 800m

5 500 m (3 NM)

Final Approach

Landing

Lidar Windprofiler

Sodar/RassT° Profiler

UHF Windprofiler

Anemometers

Lidar 3D Wind Tracer

X-bandRadar 3D

Wind Mnitoring

One Instance of Weather Sensors Deployment

Wind info

Fusion


CDG : Surveillance Area


WV Predictor

Approach/Departure : Ground wake detection, tracking & alert

WIMS(in TMA)

Data Broadcast

Data Broadcast

Pilot

WV Alarm

Approach & TowerController

WV controller HMI

Ground RadarHorizontal/Vertical scanning

of Runways

Ground RadarVertical scanning

of ILS interception Area(Entering of glide slope)

Data-Link(WV Alerts)

Limours trials 2007

Orly trials 2006 & CDG 2008

crosswind

crosswind


Wake Vortex Alerts in THALES EUROCAT-T

WVAS enables Air Traffic Control to:Separate all aircraft down to radar separation minima without jeopardizing flight safetyIncrease the airports/runways capacityAddress Arrivals and/or Departures

WVAS will ingest Wake Vortex Monitoring Sensors data for Safety Net


Acknowledgements

ADP Radar Team Jean-Paul LecorreJean-Marc ReceveauDaniel DuongGilbert Herbulot

DSNA Radar TeamAndré SimonettiAlfred HarterJean Jezequel

Direction des Aires Aéronautiques (ADP CDG) : Gérard BatistellaGuillaume Auquier


Questions

HORIZONTAL SCANNING VERTICAL SCANNING

high doppler resolution x-band wake-vortex radar : cdg airport

Documents